The goal of this project is to develop a miniaturized microfluidic thermal regulator to reversibly deactivate one or multiple areas of the neocortex through thermal regulation. This device, or cooling chip, includes indwelling microthermocouples and recording microelectrodes to monitor temperature and neural response and make online adjustments of cooling parameters to reach a desired cortical temperature. The cooling chip is being designed, assembled and tested as a multi-disciplinary collaboration between four different laboratories at the University of California Davis spanning three different departments. Although previous cooling devices have been used to reduce brain activity, the significance of our new design lies in its smaller size and the presence of indwelling electrodes/thermocouple ensemble, which will greatly expand the range of animals and experiments in which it can be used. Design criteria for the cooling chip include biocompatibility with brain tissue and a structure that accommodates the geometry of the cortical area where it is placed. Innovative soft lithography fabrication of elastomeric material (i.e., polydimethylsilane, or PDMS) offers excellent biomechanical flexibility and compliance; compact device dimensions (< 9 mm3) as well as desired heat transfer properties, which have been characterized at the tissue interface. The current prototype absorbs ~ 2 kCal/min, and produces a highly localized temperature drop from 370C to 200C within a minute. This device is highly innovative because its flexibility in size and shape allow it to be used in different animal models from rats to monkeys. A primary application for the cooling chip will be to probe cortical macrocircuitry and the specific behaviors that cortical areas generate. Further, this device can be generalized easily across a number of neuroscience disciplines for studies of sensory and motor systems as well as cognitive systems such as long-term memory (e.g., hippocampus), working memory (e.g., prefrontal cortex), and attention (e.g., parietal lobe). Its user-friendly interface with commercially available hardware and software running on a laboratory PC will make it adaptable for use in any number of laboratories. Finally, questions regarding the neural basis of complex behaviors that are currently conducted almost exclusively in non-primates can now be addressed in the more ubiquitous rodent model.